Log In Sign Up

Bayesian Inference with Certifiable Adversarial Robustness

by   Matthew Wicker, et al.

We consider adversarial training of deep neural networks through the lens of Bayesian learning, and present a principled framework for adversarial training of Bayesian Neural Networks (BNNs) with certifiable guarantees. We rely on techniques from constraint relaxation of non-convex optimisation problems and modify the standard cross-entropy error model to enforce posterior robustness to worst-case perturbations in ϵ-balls around input points. We illustrate how the resulting framework can be combined with methods commonly employed for approximate inference of BNNs. In an empirical investigation, we demonstrate that the presented approach enables training of certifiably robust models on MNIST, FashionMNIST and CIFAR-10 and can also be beneficial for uncertainty calibration. Our method is the first to directly train certifiable BNNs, thus facilitating their deployment in safety-critical applications.


page 7

page 14

page 15

page 17


Learning More Robust Features with Adversarial Training

In recent years, it has been found that neural networks can be easily fo...

Probabilistic Safety for Bayesian Neural Networks

We study probabilistic safety for Bayesian Neural Networks (BNNs) under ...

SparseVLR: A Novel Framework for Verified Locally Robust Sparse Neural Networks Search

The compute-intensive nature of neural networks (NNs) limits their deplo...

Robustness Guarantees for Bayesian Inference with Gaussian Processes

Bayesian inference and Gaussian processes are widely used in application...

Provable Robustness of Adversarial Training for Learning Halfspaces with Noise

We analyze the properties of adversarial training for learning adversari...

Robust Bayesian Learning for Reliable Wireless AI: Framework and Applications

This work takes a critical look at the application of conventional machi...

Certified Training: Small Boxes are All You Need

We propose the novel certified training method, SABR, which outperforms ...